image model
Muse Image is Meta’s first image model, built for precise edits that change only what you ask, coherent multi-turn editing, multi-reference composition, and accurate rendered text — at one cent per image.
$0.01 / image at provider cost for members — see the full price breakdown
What you need to start
Write a prompt to generate, or add reference images and describe the edit or composition.
Original results
These examples use media produced for Made. Prompts are included when the production record allows them.
“Premium editorial product photograph of an unbranded translucent tangerine running shoe floating above a brushed aluminum plinth, warm gray studio backdrop, controlled softbox light, crisp realistic materials, no text, logo, or watermark.”
“Natural editorial portrait of an adult ceramic artist in a sunlit workshop, clay dust on a dark apron, authentic expression, diffused window light, realistic skin and hand detail, no text, logo, or watermark.”
“Bold graphic poster with layered cobalt and coral paper-cut shapes and one centered white sans-serif headline reading exactly CREATE WITH INTENT, no other words, logos, or watermark.”
“Architectural photograph of a compact modern reading room with walnut shelving, one sculptural moss-green chair, cream plaster walls, soft daylight, believable scale, and straight vertical lines.”
“Overhead editorial food still life of blood oranges, dark cherries, and sparkling water on pale travertine, sharp natural texture, directional morning light, no hands, text, or logo.”
“Wide cinematic illustration of a quiet coastal observatory at blue hour, layered mist, tiny warm windows, indigo ocean, coral horizon, painterly realism, and atmospheric depth.”
These are the controls exposed by Made, which may be narrower than a provider’s full API.
Describe the subject, composition, and any exact display text in quotes.
Upload one image to edit, or several to compose from — then describe the change.
Choose from nine aspect ratios, from 21:9 banners to 9:21 portrait.
Muse keeps edits coherent across turns — refine one instruction at a time.
image
Google's latest and fastest image AI. State-of-the-art generation with up to 4K resolution and web search grounding for accurate, current visuals.
Explore model →image
Qwen Image 3 is Alibaba’s latest open image model — two major versions past the original Qwen Image — combining generation and instruction-based editing with the family’s trademark reliable text rendering.
Explore model →image
Seedream 5 Lite is the fast tier of ByteDance’s Seedream 5 image family, generating and editing photoreal images at a flat rate below Nano Banana and Seedream 4.5.
Explore model →Create scroll-stopping product photos and marketing visuals in seconds. Describe your shot, pick your style, and get professional images ready for your store, social feed, or ad campaign.
Open tool →Skip the complex menus and toolbars. Edit Images With Text Prompts. Our AI photo editor understands what you want. Type "remove the background" or "make it brighter" and watch it happen instantly.
Open tool →Upload an image, describe the target dimensions or destination, and choose whether Made should fit, crop, stretch, or pad the result.
Open tool →It is Meta’s first image generation model, hosted on fal, focused on precise instruction-based editing, multi-reference composition, and accurate text rendering.